Video Tile And Brick Partition Signaling With Inferred Tile Rows
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complex syntax structure for tile and brick partitioning in video coding standards like VVC requires suboptimal bitrate signaling, especially in Versatile Video Coding (VVC), which affects efficiency.
Innovation Solution
The method involves encoding indications of tile columns and brick heights, inferring potential tile rows, and partitioning pictures into a grid of tiles with integer coding tree units, ensuring even distribution of units across partitions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If complex syntax structure is used for tile and brick partitioning signaling, then partitioning flexibility is improved, but bitrate consumption increases
Solution Approach 1:
The partitioning syntax is segmented into multiple independent elements: tile column indications, brick height indications, and inferred tile row boundaries. This allows the encoder to signal only the necessary partitioning parameters rather than complete partitioning information, reducing bitrate while maintaining flexibility.
Solution Approach 2:
The encoder performs preliminary inference of potential tile rows based on brick row alignments before final partitioning decisions are made. This preliminary action allows the decoder to reconstruct the partitioning structure with fewer explicit signals, reducing the bitrate required for signaling complex partitioning configurations.
2Manufacturing precision
If explicit tile row boundaries are signaled, then partitioning precision is improved, but syntax complexity increases
Solution Approach 1:
The patent extracts the tile row boundary information from explicit signaling and derives it implicitly from brick row alignments. By taking out the redundant explicit boundary signals and replacing them with inference based on already-signaled brick positions, the syntax complexity is reduced while maintaining partitioning precision.
Solution Approach 2:
The decoder uses the already-signaled brick height and alignment information to self-determine tile row boundaries without requiring additional explicit signals. This self-service mechanism reduces syntax complexity while preserving the ability to achieve precise partitioning.
3Ease of operation
If more partitioning parameters are signaled, then encoding control is improved, but signaling overhead increases
Solution Approach 1:
The patent merges the signaling of tile row boundaries with the existing brick partitioning signals. By combining these functions into a unified signaling approach where brick row alignments implicitly define tile rows, the number of separate parameters to be signaled is reduced, lowering signaling overhead while maintaining encoding control.
Solution Approach 2:
The brick partitioning parameters serve multiple functions: they define both the brick structure and the tile row boundaries. This multi-functionality reduces the need for separate signaling of tile row parameters, decreasing signaling overhead while preserving comprehensive encoding control.
Data Source
AI summary
A method for video encoding or decoding comprising determining a number of units to be assigned to the partitions; indicating or inferring a number of explicitly sized partitions to be assigned; indicating sizes for or a number of units in the explicitly sized partitions; and indicating or inferring a number of evenly sized partitions to be assigned.